{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "63f5a49a",
   "metadata": {},
   "source": [
    "#Institutionalized Defense Cooperation and Regime Security#\n",
    "\n",
    "#Journal of Conflict Resolution#\n",
    "\n",
    "#Md Muhibbur Rahman#\n",
    "\n",
    "#February 2026#"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "501943ac",
   "metadata": {},
   "source": [
    "### Figure 5 – Parallel Trends"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "99b7ecd7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: pandas in c:\\users\\rmuhi\\appdata\\local\\programs\\python\\python313\\lib\\site-packages (2.3.3)\n",
      "Requirement already satisfied: statsmodels in c:\\users\\rmuhi\\appdata\\local\\programs\\python\\python313\\lib\\site-packages (0.14.4)\n",
      "Requirement already satisfied: matplotlib in c:\\users\\rmuhi\\appdata\\local\\programs\\python\\python313\\lib\\site-packages (3.10.7)\n",
      "Requirement already satisfied: pyreadstat in c:\\users\\rmuhi\\appdata\\local\\programs\\python\\python313\\lib\\site-packages (1.3.2)\n",
      "Requirement already satisfied: numpy>=1.26.0 in c:\\users\\rmuhi\\appdata\\local\\programs\\python\\python313\\lib\\site-packages (from pandas) (2.2.3)\n",
      "Requirement already satisfied: python-dateutil>=2.8.2 in c:\\users\\rmuhi\\appdata\\roaming\\python\\python313\\site-packages (from pandas) (2.9.0.post0)\n",
      "Requirement already satisfied: pytz>=2020.1 in c:\\users\\rmuhi\\appdata\\local\\programs\\python\\python313\\lib\\site-packages (from pandas) (2025.1)\n",
      "Requirement already satisfied: tzdata>=2022.7 in c:\\users\\rmuhi\\appdata\\local\\programs\\python\\python313\\lib\\site-packages (from pandas) (2025.1)\n",
      "Requirement already satisfied: scipy!=1.9.2,>=1.8 in c:\\users\\rmuhi\\appdata\\local\\programs\\python\\python313\\lib\\site-packages (from statsmodels) (1.15.2)\n",
      "Requirement already satisfied: patsy>=0.5.6 in c:\\users\\rmuhi\\appdata\\local\\programs\\python\\python313\\lib\\site-packages (from statsmodels) (1.0.1)\n",
      "Requirement already satisfied: packaging>=21.3 in c:\\users\\rmuhi\\appdata\\roaming\\python\\python313\\site-packages (from statsmodels) (24.2)\n",
      "Requirement already satisfied: contourpy>=1.0.1 in c:\\users\\rmuhi\\appdata\\local\\programs\\python\\python313\\lib\\site-packages (from matplotlib) (1.3.1)\n",
      "Requirement already satisfied: cycler>=0.10 in c:\\users\\rmuhi\\appdata\\local\\programs\\python\\python313\\lib\\site-packages (from matplotlib) (0.12.1)\n",
      "Requirement already satisfied: fonttools>=4.22.0 in c:\\users\\rmuhi\\appdata\\local\\programs\\python\\python313\\lib\\site-packages (from matplotlib) (4.57.0)\n",
      "Requirement already satisfied: kiwisolver>=1.3.1 in c:\\users\\rmuhi\\appdata\\local\\programs\\python\\python313\\lib\\site-packages (from matplotlib) (1.4.8)\n",
      "Requirement already satisfied: pillow>=8 in c:\\users\\rmuhi\\appdata\\local\\programs\\python\\python313\\lib\\site-packages (from matplotlib) (11.1.0)\n",
      "Requirement already satisfied: pyparsing>=3 in c:\\users\\rmuhi\\appdata\\local\\programs\\python\\python313\\lib\\site-packages (from matplotlib) (3.2.3)\n",
      "Requirement already satisfied: narwhals>=2.0 in c:\\users\\rmuhi\\appdata\\local\\programs\\python\\python313\\lib\\site-packages (from pyreadstat) (2.13.0)\n",
      "Requirement already satisfied: six>=1.5 in c:\\users\\rmuhi\\appdata\\roaming\\python\\python313\\site-packages (from python-dateutil>=2.8.2->pandas) (1.17.0)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "[notice] A new release of pip is available: 24.3.1 -> 26.0.1\n",
      "[notice] To update, run: python.exe -m pip install --upgrade pip\n"
     ]
    }
   ],
   "source": [
    "# Install required packages\n",
    "!pip install pandas statsmodels matplotlib pyreadstat\n",
    "\n",
    "# Import libraries and load data\n",
    "from pathlib import Path\n",
    "import pandas as pd\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import statsmodels.api as sm\n",
    "import pyreadstat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "572d9f0f",
   "metadata": {},
   "outputs": [],
   "source": [
    "file_path = Path(r\"C:\\Users\\rmuhi\\OneDrive\\Desktop\\R&R_DCA_JCR (Selective Sync Conflict)\\Replication File\\JCR_replication_data.dta\")\n",
    "\n",
    "df = pd.read_stata(file_path, convert_categoricals=False)  \n",
    "\n",
    "# Note: For replication, downdload the relevant data file in your folder and adjust the file paths above accordingly.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b2fe8ee0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create first treatment\n",
    "first_treat = df[df['dca_dum'] == 1].groupby('ccode')['year'].min().reset_index()\n",
    "first_treat.columns = ['ccode', 'first_treat_year']\n",
    "df = pd.merge(df, first_treat, on='ccode', how='left')\n",
    "df['first_treat_year'] = df['first_treat_year'].fillna(0).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "acf47e5c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create event_time\n",
    "df['event_time'] = df['year'] - df['first_treat_year']\n",
    "df.loc[df['first_treat_year'] == 0, 'event_time'] = np.nan"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3282fa69",
   "metadata": {},
   "source": [
    "Figure 5 – Parallel Trends (upper left pannel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d771c2b0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Estimate ATT for each pre-treatment event_time\n",
    "event_times = sorted(df['event_time'].dropna().unique())\n",
    "att_results = []\n",
    "\n",
    "for t in event_times:\n",
    "    if t >= 0:\n",
    "        continue  \n",
    "\n",
    "    treated = df[df['event_time'] == t]\n",
    "    control = df[(df['first_treat_year'] == 0) & (df['year'].isin(treated['year'].unique()))]\n",
    "\n",
    "    if len(treated) < 5 or len(control) < 5:\n",
    "        continue\n",
    "\n",
    "    temp = pd.concat([treated, control]).copy()\n",
    "\n",
    "    # D = 1 if this unit will be treated in future but hasn't been treated yet\n",
    "    temp['D'] = (temp['first_treat_year'] > temp['year']) & (temp['first_treat_year'] > 0)\n",
    "    temp['D'] = temp['D'].astype(int)\n",
    "\n",
    "    X = sm.add_constant(temp['D']).astype(float)\n",
    "    y = temp['any_coup_dum'].astype(float)\n",
    "\n",
    "    model = sm.OLS(y, X, missing='drop').fit()\n",
    "\n",
    "    att_results.append({\n",
    "        'event_time': t,\n",
    "        'att': model.params['D'],\n",
    "        'se': model.bse['D']\n",
    "    })\n",
    "att_df = pd.DataFrame(att_results).sort_values('event_time')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "343cf809",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x450 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Pre-treatment ATT Plot (−30 to −1), with 90% and 95% CIs styled\n",
    "pretrend_df = att_df[(att_df['event_time'] >= -30) & (att_df['event_time'] < 0)]\n",
    "\n",
    "# Extract arrays\n",
    "att = pretrend_df['att'].to_numpy()\n",
    "se = pretrend_df['se'].to_numpy()\n",
    "x = pretrend_df['event_time'].to_numpy()\n",
    "\n",
    "# Confidence bounds\n",
    "ci_low_95 = att - 1.96 * se\n",
    "ci_high_95 = att + 1.96 * se\n",
    "ci_low_90 = att - 1.645 * se\n",
    "ci_high_90 = att + 1.645 * se\n",
    "\n",
    "# Start plot\n",
    "plt.figure(figsize=(6, 4.5))\n",
    "ax = plt.gca()\n",
    "ax.set_facecolor('#f5f5f5')  \n",
    "\n",
    "# 90% CI (lightest shade, outer band)\n",
    "plt.fill_between(x, ci_low_90, ci_high_90,\n",
    "                 color='#c6dbef', alpha=0.4, label='90% CI', linewidth=0)\n",
    "\n",
    "# 95% CI (slightly darker, inner band)\n",
    "plt.fill_between(x, ci_low_95, ci_high_95,\n",
    "                 color='#6baed6', alpha=0.3, label='95% CI', linewidth=0)\n",
    "# ATT line (dark blue)\n",
    "plt.plot(x, att, color='#08306B', lw=1.5, label='ATT Estimate')\n",
    "\n",
    "# Horizontal reference line at 0\n",
    "plt.axhline(0, color='black', linestyle='--', lw=1)\n",
    "\n",
    "# Labels and layout\n",
    "plt.xlabel(\"Event time\", fontsize=11)\n",
    "plt.ylabel(\"Event Study Estimate\", fontsize=11)\n",
    "plt.title(\"Full Sample, No Covariates\", fontsize=12)\n",
    "plt.grid(True, linestyle=':', color='gray', alpha=0.4)\n",
    "plt.ylim(pretrend_df['att'].min() - 0.5, pretrend_df['att'].max() + 0.5)\n",
    "plt.tight_layout()\n",
    "plt.legend(frameon=False, loc='lower right')\n",
    "\n",
    "for spine in ['top', 'right', 'left', 'bottom']:\n",
    "    ax.spines[spine].set_visible(False)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "8a798ea1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:           any_coup_dum   R-squared:                       0.004\n",
      "Model:                            OLS   Adj. R-squared:                  0.004\n",
      "Method:                 Least Squares   F-statistic:                     21.47\n",
      "Date:                Sun, 15 Feb 2026   Prob (F-statistic):           3.68e-06\n",
      "Time:                        20:23:12   Log-Likelihood:                 2151.4\n",
      "No. Observations:                5519   AIC:                            -4299.\n",
      "Df Residuals:                    5517   BIC:                            -4286.\n",
      "Df Model:                           1                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "const          0.0384      0.003     12.024      0.000       0.032       0.045\n",
      "D             -0.0205      0.004     -4.633      0.000      -0.029      -0.012\n",
      "==============================================================================\n",
      "Omnibus:                     5907.707   Durbin-Watson:                   1.713\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):           249057.403\n",
      "Skew:                           5.720   Prob(JB):                         0.00\n",
      "Kurtosis:                      33.858   Cond. No.                         2.68\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
      "Aggregate ATT (post-treatment): ATT = -0.0205, SE = 0.0044\n"
     ]
    }
   ],
   "source": [
    "# Define post-treatment status and ever-treated dyads\n",
    "df['treated'] = df['first_treat_year'] > 0\n",
    "df['post'] = (df['year'] >= df['first_treat_year']) & (df['first_treat_year'] > 0)\n",
    "\n",
    "# ATT setup: include all dyads that are either treated-post or not yet/never treated\n",
    "agg_df = df[(df['post']) | ((df['first_treat_year'] == 0) | (df['year'] < df['first_treat_year']))].copy()\n",
    "\n",
    "# Create treatment indicator\n",
    "agg_df['D'] = (agg_df['post']) & (agg_df['treated'])\n",
    "agg_df['D'] = agg_df['D'].astype(int)\n",
    "\n",
    "# Drop missing values in DV\n",
    "agg_df = agg_df.dropna(subset=['any_coup_dum'])\n",
    "\n",
    "# Run OLS model\n",
    "X = sm.add_constant(agg_df['D']).astype(float)\n",
    "y = agg_df['any_coup_dum'].astype(float)\n",
    "model = sm.OLS(y, X).fit()\n",
    "\n",
    "# Output results\n",
    "att1 = model.params['D']\n",
    "se1 = model.bse['D']\n",
    "\n",
    "print(model.summary())\n",
    "print(f\"Aggregate ATT (post-treatment): ATT = {att1:.4f}, SE = {se1:.4f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "682140dd",
   "metadata": {},
   "source": [
    "Figure 5 – Parallel Trends (upper right pannel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b9245717",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Estimate ATT for each pre-treatment event_time (with covariates)\n",
    "covariates = ['defense', 'prev_acoup', 'polity_iv', 'lnGDPpc']\n",
    "\n",
    "event_times = sorted(df['event_time'].dropna().unique())\n",
    "att_results = []\n",
    "\n",
    "for t in event_times:\n",
    "    if t >= 0:\n",
    "        continue  \n",
    "\n",
    "    treated = df[df['event_time'] == t]\n",
    "    control = df[(df['first_treat_year'] == 0) & (df['year'].isin(treated['year'].unique()))]\n",
    "\n",
    "    if len(treated) < 5 or len(control) < 5:\n",
    "        continue\n",
    "\n",
    "    temp = pd.concat([treated, control]).copy()\n",
    "\n",
    "    # D = 1 if the unit will be treated later\n",
    "    temp['D'] = (temp['first_treat_year'] > temp['year']) & (temp['first_treat_year'] > 0)\n",
    "    temp['D'] = temp['D'].astype(int)\n",
    "\n",
    "    # Define design matrix with covariates\n",
    "    X_vars = ['D'] + covariates\n",
    "    X = sm.add_constant(temp[X_vars].astype(float))\n",
    "    y = temp['any_coup_dum'].astype(float)\n",
    "\n",
    "    model = sm.OLS(y, X, missing='drop').fit()\n",
    "\n",
    "    att_results.append({\n",
    "        'event_time': t,\n",
    "        'att': model.params['D'],\n",
    "        'se': model.bse['D']\n",
    "    })\n",
    "\n",
    "att_df = pd.DataFrame(att_results).sort_values('event_time')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d6d701d9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x450 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Pre-treatment ATT Plot (−30 to −1), with 90% and 95% CIs styled\n",
    "pretrend_df = att_df[(att_df['event_time'] >= -30) & (att_df['event_time'] < 0)]\n",
    "\n",
    "# Extract arrays\n",
    "att = pretrend_df['att'].to_numpy()\n",
    "se = pretrend_df['se'].to_numpy()\n",
    "x = pretrend_df['event_time'].to_numpy()\n",
    "\n",
    "# Confidence bounds\n",
    "ci_low_95 = att - 1.96 * se\n",
    "ci_high_95 = att + 1.96 * se\n",
    "ci_low_90 = att - 1.645 * se\n",
    "ci_high_90 = att + 1.645 * se\n",
    "\n",
    "# Start plot\n",
    "plt.figure(figsize=(6, 4.5))\n",
    "ax = plt.gca()\n",
    "ax.set_facecolor('#f5f5f5')  \n",
    "\n",
    "# 90% CI (lightest shade, outer band)\n",
    "plt.fill_between(x, ci_low_90, ci_high_90,\n",
    "                 color='#c6dbef', alpha=0.4, label='90% CI', linewidth=0)\n",
    "\n",
    "# 95% CI (slightly darker, inner band)\n",
    "plt.fill_between(x, ci_low_95, ci_high_95,\n",
    "                 color='#6baed6', alpha=0.3, label='95% CI', linewidth=0)\n",
    "# ATT line (dark blue)\n",
    "plt.plot(x, att, color='#08306B', lw=1.5, label='ATT Estimate')\n",
    "\n",
    "# Horizontal reference line at 0\n",
    "plt.axhline(0, color='black', linestyle='--', lw=1)\n",
    "\n",
    "# Labels and layout\n",
    "plt.xlabel(\"Event time\", fontsize=11)\n",
    "plt.ylabel(\"Event Study Estimate\", fontsize=11)\n",
    "plt.title(\"Full Sample, Additional Covariates\", fontsize=12)\n",
    "plt.grid(True, linestyle=':', color='gray', alpha=0.4)\n",
    "plt.ylim(pretrend_df['att'].min() - 0.5, pretrend_df['att'].max() + 0.5)\n",
    "plt.tight_layout()\n",
    "plt.legend(frameon=False, loc='lower right')\n",
    "\n",
    "for spine in ['top', 'right', 'left', 'bottom']:\n",
    "    ax.spines[spine].set_visible(False)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "4a697c81",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:           any_coup_dum   R-squared:                       0.042\n",
      "Model:                            OLS   Adj. R-squared:                  0.041\n",
      "Method:                 Least Squares   F-statistic:                     39.91\n",
      "Date:                Sun, 15 Feb 2026   Prob (F-statistic):           2.82e-40\n",
      "Time:                        20:24:59   Log-Likelihood:                 1611.2\n",
      "No. Observations:                4524   AIC:                            -3210.\n",
      "Df Residuals:                    4518   BIC:                            -3172.\n",
      "Df Model:                           5                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "const          0.0808      0.014      5.617      0.000       0.053       0.109\n",
      "D             -0.0145      0.005     -2.733      0.006      -0.025      -0.004\n",
      "defense        0.0082      0.005      1.555      0.120      -0.002       0.019\n",
      "prev_acoup     0.2421      0.026      9.170      0.000       0.190       0.294\n",
      "polity_iv     -0.0010      0.000     -2.507      0.012      -0.002      -0.000\n",
      "lnGDPpc       -0.0070      0.002     -3.887      0.000      -0.011      -0.003\n",
      "==============================================================================\n",
      "Omnibus:                     4453.999   Durbin-Watson:                   1.812\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):           139820.433\n",
      "Skew:                           5.080   Prob(JB):                         0.00\n",
      "Kurtosis:                      28.269   Cond. No.                         96.7\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
      "Aggregate ATT (post-treatment): ATT = -0.0145, SE = 0.0053\n"
     ]
    }
   ],
   "source": [
    "# Define post-treatment status and ever-treated dyads\n",
    "df['treated'] = df['first_treat_year'] > 0\n",
    "df['post'] = (df['year'] >= df['first_treat_year']) & (df['first_treat_year'] > 0)\n",
    "\n",
    "# ATT setup: include all dyads that are either treated-post or not yet/never treated\n",
    "agg_df = df[(df['post']) | ((df['first_treat_year'] == 0) | (df['year'] < df['first_treat_year']))].copy()\n",
    "\n",
    "# Create treatment indicator\n",
    "agg_df['D'] = (agg_df['post']) & (agg_df['treated'])\n",
    "agg_df['D'] = agg_df['D'].astype(int)\n",
    "\n",
    "# Drop rows with missing or infinite values in covariates or DV\n",
    "model_vars = ['D'] + covariates + ['any_coup_dum']\n",
    "agg_df = agg_df[np.isfinite(agg_df[model_vars]).all(axis=1)]\n",
    "\n",
    "# Run OLS model\n",
    "X = sm.add_constant(agg_df[['D'] + covariates].astype(float))\n",
    "y = agg_df['any_coup_dum'].astype(float)\n",
    "model = sm.OLS(y, X).fit()\n",
    "\n",
    "# Output results\n",
    "att2 = model.params['D']\n",
    "se2 = model.bse['D']\n",
    "\n",
    "print(model.summary())\n",
    "print(f\"Aggregate ATT (post-treatment): ATT = {att2:.4f}, SE = {se2:.4f}\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9bd8ddff",
   "metadata": {},
   "source": [
    "Figure 5 – Parallel Trends (bottom left pannel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "11246730",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Restrict to oecd == 0\n",
    "df = df[df['oecd'] == 0].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0030a3f6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Estimate ATT for each pre-treatment event_time\n",
    "event_times = sorted(df['event_time'].dropna().unique())\n",
    "att_results = []\n",
    "\n",
    "for t in event_times:\n",
    "    if t >= 0:\n",
    "        continue  \n",
    "\n",
    "    treated = df[df['event_time'] == t]\n",
    "    control = df[(df['first_treat_year'] == 0) & (df['year'].isin(treated['year'].unique()))]\n",
    "\n",
    "    if len(treated) < 5 or len(control) < 5:\n",
    "        continue\n",
    "\n",
    "    temp = pd.concat([treated, control]).copy()\n",
    "\n",
    "    # D = 1 if this unit will be treated in future but hasn't been treated yet\n",
    "    temp['D'] = (temp['first_treat_year'] > temp['year']) & (temp['first_treat_year'] > 0)\n",
    "    temp['D'] = temp['D'].astype(int)\n",
    "\n",
    "    X = sm.add_constant(temp['D']).astype(float)\n",
    "    y = temp['any_coup_dum'].astype(float)\n",
    "\n",
    "    model = sm.OLS(y, X, missing='drop').fit()\n",
    "\n",
    "    att_results.append({\n",
    "        'event_time': t,\n",
    "        'att': model.params['D'],\n",
    "        'se': model.bse['D']\n",
    "    })\n",
    "att_df = pd.DataFrame(att_results).sort_values('event_time')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2919a0bd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x450 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Pre-treatment ATT Plot (−30 to −1), with 90% and 95% CIs styled\n",
    "pretrend_df = att_df[(att_df['event_time'] >= -30) & (att_df['event_time'] < 0)]\n",
    "\n",
    "# Extract arrays\n",
    "att = pretrend_df['att'].to_numpy()\n",
    "se = pretrend_df['se'].to_numpy()\n",
    "x = pretrend_df['event_time'].to_numpy()\n",
    "\n",
    "# Confidence bounds\n",
    "ci_low_95 = att - 1.96 * se\n",
    "ci_high_95 = att + 1.96 * se\n",
    "ci_low_90 = att - 1.645 * se\n",
    "ci_high_90 = att + 1.645 * se\n",
    "\n",
    "# Start plot\n",
    "plt.figure(figsize=(6, 4.5))\n",
    "ax = plt.gca()\n",
    "ax.set_facecolor('#f5f5f5')  \n",
    "\n",
    "# 90% CI (lightest shade, outer band)\n",
    "plt.fill_between(x, ci_low_90, ci_high_90,\n",
    "                 color='#c6dbef', alpha=0.4, label='90% CI', linewidth=0)\n",
    "\n",
    "# 95% CI (slightly darker, inner band)\n",
    "plt.fill_between(x, ci_low_95, ci_high_95,\n",
    "                 color='#6baed6', alpha=0.3, label='95% CI', linewidth=0)\n",
    "# ATT line (dark blue)\n",
    "plt.plot(x, att, color='#08306B', lw=1.5, label='ATT Estimate')\n",
    "\n",
    "# Horizontal reference line at 0\n",
    "plt.axhline(0, color='black', linestyle='--', lw=1)\n",
    "\n",
    "# Labels and layout\n",
    "plt.xlabel(\"Event time\", fontsize=11)\n",
    "plt.ylabel(\"Event Study Estimate\", fontsize=11)\n",
    "plt.title(\"Developing Country Sample, No Covariates\", fontsize=12)\n",
    "plt.grid(True, linestyle=':', color='gray', alpha=0.4)\n",
    "plt.ylim(pretrend_df['att'].min() - 0.5, pretrend_df['att'].max() + 0.5)\n",
    "plt.tight_layout()\n",
    "plt.legend(frameon=False, loc='lower right')\n",
    "\n",
    "for spine in ['top', 'right', 'left', 'bottom']:\n",
    "    ax.spines[spine].set_visible(False)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "48dd0c34",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:           any_coup_dum   R-squared:                       0.003\n",
      "Model:                            OLS   Adj. R-squared:                  0.003\n",
      "Method:                 Least Squares   F-statistic:                     13.64\n",
      "Date:                Sun, 15 Feb 2026   Prob (F-statistic):           0.000224\n",
      "Time:                        20:26:59   Log-Likelihood:                 1479.6\n",
      "No. Observations:                4682   AIC:                            -2955.\n",
      "Df Residuals:                    4680   BIC:                            -2942.\n",
      "Df Model:                           1                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "const          0.0414      0.004     11.586      0.000       0.034       0.048\n",
      "D             -0.0191      0.005     -3.693      0.000      -0.029      -0.009\n",
      "==============================================================================\n",
      "Omnibus:                     4717.149   Durbin-Watson:                   1.714\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):           152421.628\n",
      "Skew:                           5.272   Prob(JB):                         0.00\n",
      "Kurtosis:                      28.887   Cond. No.                         2.57\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
      "Aggregate ATT (post-treatment): ATT = -0.0205, SE = 0.0044\n"
     ]
    }
   ],
   "source": [
    "# Define post-treatment status and ever-treated dyads\n",
    "df['treated'] = df['first_treat_year'] > 0\n",
    "df['post'] = (df['year'] >= df['first_treat_year']) & (df['first_treat_year'] > 0)\n",
    "\n",
    "# ATT setup: include all dyads that are either treated-post or not yet/never treated\n",
    "agg_df = df[(df['post']) | ((df['first_treat_year'] == 0) | (df['year'] < df['first_treat_year']))].copy()\n",
    "\n",
    "# Create treatment indicator\n",
    "agg_df['D'] = (agg_df['post']) & (agg_df['treated'])\n",
    "agg_df['D'] = agg_df['D'].astype(int)\n",
    "\n",
    "# Drop missing values in DV\n",
    "agg_df = agg_df.dropna(subset=['any_coup_dum'])\n",
    "\n",
    "# Run OLS model\n",
    "X = sm.add_constant(agg_df['D']).astype(float)\n",
    "y = agg_df['any_coup_dum'].astype(float)\n",
    "model = sm.OLS(y, X).fit()\n",
    "\n",
    "# Output results\n",
    "att3 = model.params['D']\n",
    "se3 = model.bse['D']\n",
    "\n",
    "print(model.summary())\n",
    "print(f\"Aggregate ATT (post-treatment): ATT = {att1:.4f}, SE = {se1:.4f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ba58a7f",
   "metadata": {},
   "source": [
    "Figure 5 – Parallel Trends (bottom right pannel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c0f9494b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Estimate ATT for each pre-treatment event_time (with covariates)\n",
    "covariates = ['defense', 'prev_acoup', 'polity_iv', 'lnGDPpc']\n",
    "\n",
    "event_times = sorted(df['event_time'].dropna().unique())\n",
    "att_results = []\n",
    "\n",
    "for t in event_times:\n",
    "    if t >= 0:\n",
    "        continue  \n",
    "\n",
    "    treated = df[df['event_time'] == t]\n",
    "    control = df[(df['first_treat_year'] == 0) & (df['year'].isin(treated['year'].unique()))]\n",
    "\n",
    "    if len(treated) < 5 or len(control) < 5:\n",
    "        continue\n",
    "\n",
    "    temp = pd.concat([treated, control]).copy()\n",
    "\n",
    "    # D = 1 if the unit will be treated later\n",
    "    temp['D'] = (temp['first_treat_year'] > temp['year']) & (temp['first_treat_year'] > 0)\n",
    "    temp['D'] = temp['D'].astype(int)\n",
    "\n",
    "    # Define design matrix with covariates\n",
    "    X_vars = ['D'] + covariates\n",
    "    X = sm.add_constant(temp[X_vars].astype(float))\n",
    "    y = temp['any_coup_dum'].astype(float)\n",
    "\n",
    "    model = sm.OLS(y, X, missing='drop').fit()\n",
    "\n",
    "    att_results.append({\n",
    "        'event_time': t,\n",
    "        'att': model.params['D'],\n",
    "        'se': model.bse['D']\n",
    "    })\n",
    "\n",
    "att_df = pd.DataFrame(att_results).sort_values('event_time')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2893958d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x450 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Pre-treatment ATT Plot (−30 to −1), with 90% and 95% CIs styled\n",
    "pretrend_df = att_df[(att_df['event_time'] >= -30) & (att_df['event_time'] < 0)]\n",
    "\n",
    "# Extract arrays\n",
    "att = pretrend_df['att'].to_numpy()\n",
    "se = pretrend_df['se'].to_numpy()\n",
    "x = pretrend_df['event_time'].to_numpy()\n",
    "\n",
    "# Confidence bounds\n",
    "ci_low_95 = att - 1.96 * se\n",
    "ci_high_95 = att + 1.96 * se\n",
    "ci_low_90 = att - 1.645 * se\n",
    "ci_high_90 = att + 1.645 * se\n",
    "\n",
    "# Start plot\n",
    "plt.figure(figsize=(6, 4.5))\n",
    "ax = plt.gca()\n",
    "ax.set_facecolor('#f5f5f5')  \n",
    "\n",
    "# 90% CI (lightest shade, outer band)\n",
    "plt.fill_between(x, ci_low_90, ci_high_90,\n",
    "                 color='#c6dbef', alpha=0.4, label='90% CI', linewidth=0)\n",
    "\n",
    "# 95% CI (slightly darker, inner band)\n",
    "plt.fill_between(x, ci_low_95, ci_high_95,\n",
    "                 color='#6baed6', alpha=0.3, label='95% CI', linewidth=0)\n",
    "# ATT line (dark blue)\n",
    "plt.plot(x, att, color='#08306B', lw=1.5, label='ATT Estimate')\n",
    "\n",
    "# Horizontal reference line at 0\n",
    "plt.axhline(0, color='black', linestyle='--', lw=1)\n",
    "\n",
    "# Labels and layout\n",
    "plt.xlabel(\"Event time\", fontsize=11)\n",
    "plt.ylabel(\"Event Study Estimate\", fontsize=11)\n",
    "plt.title(\"Developing Country Sample, Additional Covariates\", fontsize=12)\n",
    "plt.grid(True, linestyle=':', color='gray', alpha=0.4)\n",
    "plt.ylim(pretrend_df['att'].min() - 0.5, pretrend_df['att'].max() + 0.5)\n",
    "plt.tight_layout()\n",
    "plt.legend(frameon=False, loc='lower right')\n",
    "\n",
    "for spine in ['top', 'right', 'left', 'bottom']:\n",
    "    ax.spines[spine].set_visible(False)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "f4ed7c95",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:           any_coup_dum   R-squared:                       0.038\n",
      "Model:                            OLS   Adj. R-squared:                  0.037\n",
      "Method:                 Least Squares   F-statistic:                     29.46\n",
      "Date:                Sun, 15 Feb 2026   Prob (F-statistic):           1.95e-29\n",
      "Time:                        20:28:17   Log-Likelihood:                 990.74\n",
      "No. Observations:                3719   AIC:                            -1969.\n",
      "Df Residuals:                    3713   BIC:                            -1932.\n",
      "Df Model:                           5                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "const          0.0985      0.019      5.294      0.000       0.062       0.135\n",
      "D             -0.0176      0.006     -2.831      0.005      -0.030      -0.005\n",
      "defense        0.0088      0.006      1.395      0.163      -0.004       0.021\n",
      "prev_acoup     0.2405      0.029      8.233      0.000       0.183       0.298\n",
      "polity_iv     -0.0011      0.000     -2.499      0.012      -0.002      -0.000\n",
      "lnGDPpc       -0.0095      0.002     -3.900      0.000      -0.014      -0.005\n",
      "==============================================================================\n",
      "Omnibus:                     3398.157   Durbin-Watson:                   1.809\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):            78348.467\n",
      "Skew:                           4.624   Prob(JB):                         0.00\n",
      "Kurtosis:                      23.496   Cond. No.                         75.4\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
      "Aggregate ATT (post-treatment): ATT = -0.0145, SE = 0.0053\n"
     ]
    }
   ],
   "source": [
    "# Define post-treatment status and ever-treated dyads\n",
    "df['treated'] = df['first_treat_year'] > 0\n",
    "df['post'] = (df['year'] >= df['first_treat_year']) & (df['first_treat_year'] > 0)\n",
    "\n",
    "# ATT setup: include all dyads that are either treated-post or not yet/never treated\n",
    "agg_df = df[(df['post']) | ((df['first_treat_year'] == 0) | (df['year'] < df['first_treat_year']))].copy()\n",
    "\n",
    "# Create treatment indicator\n",
    "agg_df['D'] = (agg_df['post']) & (agg_df['treated'])\n",
    "agg_df['D'] = agg_df['D'].astype(int)\n",
    "\n",
    "# Drop rows with missing or infinite values in covariates or DV\n",
    "model_vars = ['D'] + covariates + ['any_coup_dum']\n",
    "agg_df = agg_df[np.isfinite(agg_df[model_vars]).all(axis=1)]\n",
    "\n",
    "# Run OLS model\n",
    "X = sm.add_constant(agg_df[['D'] + covariates].astype(float))\n",
    "y = agg_df['any_coup_dum'].astype(float)\n",
    "model = sm.OLS(y, X).fit()\n",
    "\n",
    "# Output results\n",
    "att4 = model.params['D']\n",
    "se4 = model.bse['D']\n",
    "\n",
    "print(model.summary())\n",
    "print(f\"Aggregate ATT (post-treatment): ATT = {att2:.4f}, SE = {se2:.4f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "24a9f8aa",
   "metadata": {},
   "source": [
    "### Figure 6– Average Treatment Effects on the Treated"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "83a93fb0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x320 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "\n",
    "# Step 1: Prepare the data\n",
    "group1 = pd.DataFrame({\n",
    "    'Model': ['Full Sample', 'Developing Country Sample'],\n",
    "    'att': [att1, att3],\n",
    "    'se': [se1, se3]\n",
    "})\n",
    "group1['ci_low'] = group1['att'] - 1.96 * group1['se']\n",
    "group1['ci_high'] = group1['att'] + 1.96 * group1['se']\n",
    "\n",
    "group2 = pd.DataFrame({\n",
    "    'Model': ['Full Sample', 'Developing Country Sample'],\n",
    "    'att': [att2, att4],\n",
    "    'se': [se2, se4]\n",
    "})\n",
    "group2['ci_low'] = group2['att'] - 1.96 * group2['se']\n",
    "group2['ci_high'] = group2['att'] + 1.96 * group2['se']\n",
    "\n",
    "# Step 2: Positions\n",
    "y_positions = [1, 0]\n",
    "dot_color = '#e34a33'\n",
    "\n",
    "# Step 3: Subplots\n",
    "fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(10, 3.2), sharey=True)\n",
    "\n",
    "# === Left plot: No Covariates ===\n",
    "ax1.set_facecolor('#f9f9f9')\n",
    "for spine in ['top', 'right', 'left', 'bottom']:\n",
    "    ax1.spines[spine].set_visible(False)\n",
    "\n",
    "ax1.errorbar(group1['att'], y_positions,\n",
    "             xerr=1.96 * group1['se'],\n",
    "             fmt='o', color=dot_color, ecolor=dot_color,\n",
    "             elinewidth=1, capsize=2, markersize=6)\n",
    "\n",
    "ax1.axvline(0, color='black', linestyle='--', linewidth=1)\n",
    "ax1.set_xlabel(\"ATT\", fontsize=10)\n",
    "ax1.set_title(\"No Covariates\", fontsize=11)\n",
    "ax1.set_xlim(group1['ci_low'].min() - 0.02, group1['ci_high'].max() + 0.025)\n",
    "ax1.set_ylim(-0.5, 1.5)\n",
    "ax1.set_yticks(y_positions)\n",
    "ax1.set_yticklabels(group1['Model'], fontsize=10)\n",
    "ax1.grid(True, linestyle=':', linewidth=0.7, alpha=0.6)\n",
    "\n",
    "# === Right plot: With Covariates ===\n",
    "ax2.set_facecolor('#f9f9f9')\n",
    "for spine in ['top', 'right', 'left', 'bottom']:\n",
    "    ax2.spines[spine].set_visible(False)\n",
    "\n",
    "ax2.errorbar(group2['att'], y_positions,\n",
    "             xerr=1.96 * group2['se'],\n",
    "             fmt='o', color=dot_color, ecolor=dot_color,\n",
    "             elinewidth=1, capsize=2, markersize=6)\n",
    "\n",
    "ax2.axvline(0, color='black', linestyle='--', linewidth=1)\n",
    "ax2.set_xlabel(\"ATT\", fontsize=10)\n",
    "ax2.set_title(\"Additional Covariates\", fontsize=11)\n",
    "ax2.set_xlim(group2['ci_low'].min() - 0.02, group2['ci_high'].max() + 0.02)\n",
    "ax2.set_ylim(-0.5, 1.5)\n",
    "# Don't set y-ticklabels again on ax2\n",
    "ax2.grid(True, linestyle=':', linewidth=0.7, alpha=0.6)\n",
    "\n",
    "# Adjust spacing to show y-axis labels well\n",
    "plt.subplots_adjust(left=0.15, right=0.95, wspace=0.25)\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ca5a2276",
   "metadata": {},
   "source": [
    "#*****************#"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.13.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}
